Toggle light / dark theme

A machine-learning algorithm demonstrated the capability to process data that exceeds a computer’s available memory by identifying a massive data set’s key features and dividing them into manageable batches that don’t choke computer hardware. Developed at Los Alamos National Laboratory, the algorithm set a world record for factorizing huge data sets during a test run on Oak Ridge National Laboratory’s Summit, the world’s fifth-fastest supercomputer.

Equally efficient on laptops and supercomputers, the highly scalable solves hardware bottlenecks that prevent processing information from data-rich applications in , , social media networks, national security science and earthquake research, to name just a few.

“We developed an ‘out-of-memory’ implementation of the non-negative matrix factorization method that allows you to factorize larger than previously possible on a given hardware,” said Ismael Boureima, a computational physicist at Los Alamos National Laboratory. Boureima is first author of the paper in The Journal of Supercomputing on the record-breaking algorithm.

The ultimate answer to the supply chain issue.


A school project for motion graphics, put together several of the product advertisements found in Philip K Dick’s 1969 sci-fi novel into a full 1m commercial. Aimed for surrealism.

The universe is big, as Douglas Adams would say.

The most distant light we can see is the (CMB), which has taken more than 13 billion years to reach us. This marks the edge of the , and while you might think that means the is 26 billion light-years across, thanks to it is now closer to 46 billion light-years across. By any measure, this is pretty darn big. But most cosmologists think the universe is much larger than our observable corner of it. That what we can see is a small part of an unimaginably vast, if not infinite creation. However, a new paper published on the arXiv preprint server argues that the observable universe is mostly all there is.

In other words, on a cosmic scale, the universe is quite small.

This talk is about how you can use wireless signals and fuse them with vision and other sensing modalities through AI algorithms to give humans and robots X-ray vision to see objects hidden inside boxes or behind other object.

Tara Boroushaki is a Ph.D student at MIT. Her research focuses on fusing radio frequency (RF) sensing with vision through artificial intelligence. She designs algorithms and builds systems that leverage such fusion to enable capabilities that were not feasible before in applications spanning augmented reality, virtual reality, robotics, smart homes, and smart manufacturing. This talk was given at a TEDx event using the TED conference format but independently organized by a local community.

Congestive heart failure (also called heart failure) is a serious condition in which the heart doesn’t pump blood as efficiently as it should. Despite its name, heart failure doesn’t mean that the heart has literally failed or is about to stop working. Rather, it means that the heart muscle has become less able to contract over time or has a mechanical problem that limits its ability to fill with blood. As a result, it can’t keep up with the body’s demand, and blood returns to the heart faster than it can be pumped out—it becomes congested, or backed up. This pumping problem means that not enough oxygen-rich blood can get to the body’s other organs.

The body tries to compensate in different ways. The heart beats faster to take less time for refilling after it contracts—but over the long run, less blood circulates, and the extra effort can cause heart palpitations. The heart also enlarges a bit to make room for the blood. The lungs fill with fluid, causing shortness of breath. The kidneys, when they don’t receive enough blood, begin to retain water and sodium, which can lead to kidney failure. With or without treatment, heart failure is often and typically progressive, meaning it gradually gets worse.

More than 5 million people in the United States have congestive heart failure. It’s the most common diagnosis in hospitalized patients over age 65. One in nine deaths has heart failure as a contributing cause.

We don’t treat Isla’s vision loss as a sad circumstance or as something that is broken in her. It’s so important for us that she knows her vision impairment is not something that makes her less than. If anything, it makes her a stronger, more amazing person, and we couldn’t be prouder of who she is.

Vision impairment is the only symptom she displays of this disease, and we are fighting with everything we have to ensure it stays this way. We were told on diagnosis day that that day was the healthiest Isla would ever be, and that she was at her peak; two years later, and she has continued to defy that, Stockdale added.

The family believes that Isla’s incredible development is a result of the medication, called Miglustat, which she has been taking since November 2022.

Testing time: To see whether the modified VR headset could accurately record brain activity, the researchers had four participants wear it while opening or closing their eyes on command. There is typically an increase in alpha waves in a person’s brain when their eyes are closed, and the researchers could see this increase in the EEG data recorded by their headset.

They also had the participants wear the headset while playing a custom game designed to trigger a type of brain activity that occurs when a person is preparing for an event. As hoped, the researchers were able to see this activity in the recorded EEG data, too.

Looking ahead: The UT Austin team has started filing the paperwork to patent their tech and say they’d be open to working with VR companies to incorporate it into their systems. They don’t say what it costs to make the soft electrodes, so it’s not clear exactly how much consumers or professionals should expect to pay for a modified VR headset.